Studies on the Collective Species Viola 

 tricolor L. 



(Preliminary Notes.) 

 By 



J. Clausen. 



With Plates I— III. 



In the summer of 1918 during my military service I began 

 a study of the collective species Viola tricolor L. (I use this name 

 in its most comprehensive sense to include both Viola tricolor 

 L. and V. arvensis Murr, with all their sub-forms). The investi- 

 gation is planned to comprise: 



1) Decomposition of the species into elementary species. 



2) The relationship to other species of the Melanium section of 

 the genus Viola. 



3) Cytological investigation and 



4) The problems of heredity. 



I. Systematic Studies. 



Viola tricolor has long ago been divided by a multitude of 

 authors into many small systematic units. Wittrock (1897), in 

 his excellent treatise, deals with about 40 sub-species, forms, 

 sub-forms and varieties. Becker (1910) has about 20 sub-species. 



In order to estimate the extent to which the group can be 

 divided, I resolved, — being partly influenced by Raunkiær's 

 interpretation of the Isoreagent as the Elementary Species 

 (Raunkiær 1918) — to work as if all the old sub-divisions did 

 not exist, only taking in consideration the characters and their 

 variations alone. As the number of the varying characters is very 

 large, I was obliged to limit the examination only to a few. In 

 chosing those, I derived great help from Wittrock's treatise 

 (1897, p. 51—55; 80—81). 



